We conducted a rapid overview of systematic reviews (SRs) of evidence on the effectiveness of MLHPs within a span of about eight weeks and in all domains corresponding to the CPHC package in Ayushman Bharat. The 12 CPHC packages are: pregnancy and childbirth; neonatal and infant health services; childhood and adolescent health services; family planning, contraceptive services and other reproductive care services; communicable diseases (prevention and management); non-communicable diseases; elderly and palliative care; oral health care; ophthalmic and ear, nose and throat (ENT) care; mental health and emergency medical services ³ .

The World Health Organization (WHO), defined MLHP as “a health provider who is trained, authorised and regulated to work autonomously, receives pre-service training at a higher education institution for at least 2–3 years and whose scope of practice includes (but is not restricted to) being able to diagnose, manage and treat illness, disease, and impairments (including performing surgery, where appropriately trained), prescribe medicines, as well as engage in preventive and promotive care” ^{1, p.8}. However, MLHPs in various countries have been variously referred to as substitute health workers, auxiliaries, non-physician clinicians, and include cadres such as clinical officers, medical assistants, physician assistants, nurse practitioners, and surgical technicians. Institutions and researchers worldwide use alternate or less well-specified definitions, and therefore MLHP as defined in the SRs was considered for this review. Therefore, we used broad criteria for the rapid overview wherein we accepted the definition of MLHPs as defined by the SR authors. The overview of SRs is an appropriate study design for our research because we intended to summarise the evidence for multiple conditions in different disease/condition domains for the same type of intervention and on similar health systems, clinical and public health outcomes.

Participants. The RES considered SRs assessing outcomes for participants receiving care from MLHPs in LMICs, including India.

Intervention and comparators. SRs that compared service delivery provided by MLHPs with doctors or other types of MLHPs were included. The MLHPs included were midwives, nurses, auxiliary nurses, nurse assistants, non‐physician clinicians, and surgical technicians.

Outcomes. The following outcomes were considered for inclusion based on the initial discussions with the requester: healthcare and clinical outcomes (mortality, morbidity, outcomes associated with care delivery, and physiological measures); access to care; and quality of care (including patient or client satisfaction with care).

Study design. SRs including studies of any quantitative study design, irrespective of whether they have or have not conducted meta-analyses and irrespective of whether they have or have not used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework to assess the certainty of evidence were included. Qualitative SRs were not considered.

Context. SRs with a focus on and/or including studies from LMICs and India were included.

Given time constraints, the search was limited to published and indexed articles, and those published in the English language. The following databases were searched (from database inception up until March 2019): Cochrane Database of Systematic Reviews; Medline (PubMed); EMBASE; Health Systems Evidence; and CINAHL. An additional search was conducted from April 2019 to April 2020 to update the review findings for recency and relevancy. Search strategies (for both the periods) are provided separately for each database (see Extended data) ⁵ .

The lead reviewer (SM) independently screened the titles and abstracts of studies for inclusion, following which full-text examination of eligible studies was conducted for potential inclusion. A second reviewer (SB) randomly verified the results of the study selection process during both the screening stages. For each domain of interest, where multiple SRs were available, only one SR was included based on its comprehensiveness, recency, and quality. Each SR was independently assessed for methodological quality by using established standardised criteria (A MeaSurement Tool to Assess systematic Reviews (AMSTAR) 2 checklist) ⁶ . Data from included reviews was extracted using a pre-defined template, which included variables such as review type, review question, countries/settings, participants characteristics, interventions, outcome measures and review conclusions. The lead author (SM) independently extracted all relevant outcome data, with random verification of 20% of the included studies by another author (SB).

The GRADE approach was used to assess the certainty of the evidence using a transparent framework for developing and presenting the summary of findings tables ^{7– 10} . The GRADE of evidence was synthesised with respect to a PHC setting and in an LMIC context to make the product locally relevant ^{9, 10} .

As part of the RES process, the RES team and NHSRC jointly convened a policy dialogue to engage and consult with relevant stakeholders to present an interim draft of the MLHP policy brief. The stakeholders included policy makers (key stakeholders from government agencies and collaborators), health system managers, and researchers from more than eight states in India.

The search for evidence identified 5171 studies ( Figure 1 – Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram). Following the study screening process, full-text articles were retrieved for 30 potentially relevant studies. In cases where there were multiple SRs for the same domain, the SR that was the most recent and provided comprehensive information (as per authors’ (SM, SB) consensus) was selected and included. Following full-text examination, 23 out of 30 SRs were excluded. An additional 717 records were identified in an updated search. However, following the study selection process, none of the reviews were found to be relevant to the topic of interest ( Figure 2 – PRISMA flow diagram (updated search)). Overall, seven SRs were included in the RES.

The majority of the studies included in the SRs were randomised controlled trials (RCTs), with some quasi-experimental study designs and observational studies. Key characteristics of the included SRs are provided in the Extended data file ⁵ . All SRs, except one ¹¹ included studies that were mostly conducted in LMICs. The studies related to HIV/AIDS were mostly conducted in sub-Saharan African countries. Most studies compared care provided by midwives or auxiliary nurse midwives or nurses with that provided by doctors working in a team along with midwives or nurses.

The AMSTAR-2 checklist ⁶ was used to assess the methodological quality of SRs included in the report. The checklist is a 16-item questionnaire. The critical appraisal results of the included SRs are provided in the Extended data file ⁵ . Six out of seven SRs were of moderate to high methodological quality and well reported. Almost all the SRs did not refer to a priori protocol and publication bias was not assessed. One SR by Chaudhary et al. was of poor quality, as assessed by the checklist ¹² .

Key findings. The key findings from the included SRs have been categorised based on the various healthcare domains of interest in the CPHC package ³ . The quality of evidence for the main outcomes is summarised using the GRADE approach and ‘Summary of Findings’ tables ^{7– 10} . The Summary of Findings tables aid in recording results, outcomes, and outcome risks in a structured synthesis format.

MLHPs for care in pregnancy and childbirth. An SR compared the effectiveness of care provided by MLHPs, particularly midwives and auxiliary nurse midwives with doctors providing care in a team with midwives ¹¹ . The review included patients receiving pregnancy and childbirth services including antenatal care. The majority of the studies were conducted in tertiary care settings and developed countries. Most of the evidence was assessed as low certainty. It was found that the use of intrapartum analgesia and episiotomies were less likely with care provided by midwives when compared with that provided by doctors working along with midwives. Also, no significant difference in rates for performing caesarean section, postpartum haemorrhage, and preterm births were reported.

No significant difference in the likelihood of an incomplete abortion was reported between groups of patients treated by auxiliary nurse midwives compared to those cared for by doctors. However, the likelihood of a complication during or an adverse event after manual vacuum aspiration was significantly greater with care provided by auxiliary nurse midwives. There was very low certainty evidence to suggest that pregnancy care provided by clinical officers reduced the likelihood of early neonatal death or postoperative maternal health outcomes, such as fever and wound infections. Table 1 provides a summary of findings and certainty of evidence related to pregnancy and childbirth care provided by midwives, auxiliary nurse midwives and clinical officers with that provided by doctors in a team with midwives.

MLHPs for neonatal and infant health care services. The effectiveness of midwives/nurses delivering care for neonatal and infant healthcare services was compared with that provided by doctors or obstetricians in a team with midwives in a SR ¹¹ . The population included patients receiving neonatal and infant health services. The majority of the studies were conducted in tertiary care settings and developed countries. The certainty of the evidence was assessed as low quality. The review results showed that there was no significant difference between the groups in foetal or neonatal death rates. None of the studies included in the review reported on clinical outcomes, and outcomes related to quality of care and access to care. Table 2 presents the review findings in plain language format and the certainty of the evidence for the relevant outcome.

MLHPs for family planning, contraceptive and other reproductive health care services. Another SR by Barnard et al. evaluated the safety and effectiveness of surgical and medical abortion procedures administered by MLHPs compared to doctors ¹³ . The review included various MLHPs who included nurses, midwives, doctor assistants, and physician assistants delivering care for patients requesting abortion procedures, either surgical or medical. The majority of the studies were conducted in PHC settings and LMICs. Much of the evidence was of low or very low quality. The review found that the evidence for surgical abortion procedures provided by MLHPs was lacking. Further, evidence from cohort studies suggested that there was an increase in the risk of failure or incomplete abortion for surgical abortion procedures when provided by MLHPs. However, no statistically significant differences in complications alone, immediate complications or delayed complications were reported when surgical abortion was provided by MLHPs. Concerning medical abortion procedures, the review results suggested MLHPs could safely and effectively carry out these procedures. No significant differences were reported for abortion failure or incomplete abortion. None of the studies included in the SR examined other outcomes of interest such as mortality, quality of care, and access to care. Table 3 presents a summary of findings on various outcomes related to surgical and medical abortion procedures provided by MLHPs compared to doctors.

MLHPs for communicable diseases. Two SRs examined the effectiveness of the delivery of antiretroviral therapy (ART) provided by MLHPs in HIV-infected patients ^{14, 12} . The reviews included studies mainly conducted in primary health care settings and LMICs. The studies included in the reviews compared ART provided by nurses or clinical officers with doctors. The certainty of the evidence varied for different outcomes, from high to very low quality. However, the evidence for various outcomes was based on relatively few studies. The review reported that there was no significant difference in mortality, with lower rates of losses to follow up at 12 months. Further, no difference in death or number of patients lost to follow up at 12 months was reported when doctors initiated therapy and nurses provided follow-up. The reviews suggested that shifting tasks from doctors to MLHPs may help in potentially reducing costs of ART provision, without compromising on the quality of care and patient outcomes. Table 4 provides a summary of findings reported in the SRs for outcomes related to the initiation and maintenance of ART in HIV-infected patients.

MLHPs for non-communicable diseases. Two reviews compared the effectiveness of care provided by non-physician health workers (NPHWs) for patients with non-communicable diseases in primary and secondary health care settings ^{15, 16} . The NPHWs included nurses, pharmacists, allied health professionals, and physician assistants. The care provided by NPHWs was compared to that provided by doctors for various physiological measure outcomes, health-related quality of life, and access to care. The evidence assessed was of moderate to high quality. The findings from the two reviews suggested that care provided by NPHWs with varying but high degrees of autonomy and with support was comparable to that provided by doctors for various relevant outcomes. Care prescription by NPHWs significantly improved outcomes such as systolic blood pressure, glycated haemoglobin and low-density lipoprotein levels. Also, the care provided by NPHWs improved health-related quality of life (physical component). However, the mental health-related quality of life was reduced with the care provided by NPHWs compared to that provided by doctors. There was a lack of conclusive evidence on outcomes related to access to care. Table 5 presents a summary of findings for various relevant outcomes related to chronic diseases.

MLHPs for mental health. One SR compared the effectiveness of delivery of care provided by non-specialist health workers (NSHWs) to that provided by mental health specialists in women with perinatal depression ¹⁷ . The NSHWs included midwives, nurses, and community health workers. The studies included in the review were conducted in primary health settings and LMICs. The review found that the NSHWs could effectively deliver psychological interventions for perinatal depression in low-resource settings, particularly where specialist services are both scarce and expensive. The review did not examine other relevant outcomes such as mortality, quality of care, and access to care. The review lacked proper reporting and hence it was not possible to assess the certainty of evidence by GRADE. The SR included nine RCTs involving a total of 14,555 participants. Table 6 briefly presents a narrative summary of the findings reported in the review.

MLHPs for other packages of care. This RES did not identify any SRs that assessed the role of MLHPs in the provision of following health services.

All data underlying the results are available as part of the article and no additional source data are required.

Figshare: Extended data.docx. https://doi.org/10.6084/m9.figshare.12401525.v1 ⁵

This project contains the following extended data:

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).